The present invention is directed to integrated circuits. More particularly, the invention provides a system and method for current regulation. Merely by way of example, the invention has been applied to power conversion systems in quasi-resonance mode. But it would be recognized that the invention has a much broader range of applicability.
Light emitting diodes (LEDs) are widely used for lighting applications. Oftentimes, approximately constant currents are used to control working currents of LEDs to achieve constant brightness. FIG. 1 is a simplified diagram showing a conventional power conversation system for LED lighting. The power conversion system 100 includes a controller 102, resistors 104, 124, 126 and 132, capacitors 106, 120 and 134, a diode 108, a transformer 110 including a primary winding 112, a secondary winding 114 and an auxiliary winding 116, a power switch 128, a current sensing resistor 130, and a rectifying diode 118. The controller 102 includes terminals (e.g., pins) 138, 140, 142, 144, 146 and 148. For example, the power switch 128 is a bipolar junction transistor. In another example, the power switch 128 is a MOS transistor.
An alternate-current (AC) input voltage 152 is applied to the system 100. A bulk voltage 150 (e.g., a rectified voltage no smaller than 0 V) associated with the AC input voltage 152 is received by the resistor 104. The capacitor 106 is charged in response to the bulk voltage 150, and a voltage 154 is provided to the controller 102 at the terminal 138 (e.g., terminal VCC). If the voltage 154 is larger than a predetermined threshold voltage in magnitude, the controller 102 begins to operate normally, and outputs a drive signal 156 through the terminal 142 (e.g., terminal GATE). For example, the drive signal 156 is a pulse-width-modulation (PWM) signal with a switching frequency and a duty cycle. The switch 128 is closed (e.g., being turned on) or open (e.g., being turned off) in response to the drive signal 156 so that the output current 158 is regulated to be approximately constant.
The auxiliary winding 116 charges the capacitor 106 through the diode 108 when the switch 128 is opened (e.g., being turned off) in response to the drive signal 156 so that the controller 102 can operate normally. For example, a feedback signal 160 is provided to the controller 102 through the terminal 140 (e.g., terminal FB) in order to detect the end of a demagnetization process of the secondary winding 118 for charging or discharging the capacitor 134 using an internal error amplifier in the controller 102. In another example, the feedback signal 160 is provided to the controller 102 through the terminal 140 (e.g., terminal FB) in order to detect the beginning and the end of the demagnetization process of the secondary winding 118. The resistor 130 is used for detecting a primary current 162 flowing through the primary winding 112, and a current-sensing signal 164 is provided to the controller 102 through the terminal 144 (e.g., terminal CS) to be processed during each switching cycle. Peak magnitudes of the current-sensing signal 164 are sampled and provided to the internal error amplifier. The capacitor 120 is used to maintain an output voltage 168 so as to keep a stable output current through an output load (e.g., one or more LEDs 122). For example, the system 100 operates in a quasi-resonant mode.
FIG. 2 is a simplified conventional diagram showing the controller 102 as part of the system 100. The controller 102 includes a ramp-signal generator 202, an under-voltage lock-out (UVLO) component 204, a modulation component 206, a logic controller 208, a driving component 210, a demagnetization detector 212, an error amplifier 216, and a current-sensing component 214.
As shown in FIG. 2, the UVLO component 204 detects the signal 154 and outputs a signal 218. If the signal 154 is larger than a first predetermined threshold in magnitude, the controller 102 begins to operate normally. If the signal 154 is smaller than a second predetermined threshold in magnitude, the controller 102 is turned off. The second predetermined threshold is smaller than the first predetermined threshold in magnitude. The error amplifier 216 receives a signal 220 from the current-sensing component 214 and a reference signal 222 and outputs an amplified signal 224 to the modulation component 206. The modulation component 206 also receives a signal 228 from the ramp-signal generator 202 and outputs a modulation signal 226. For example, the signal 228 is a ramping signal and increases, linearly or non-linearly, to a peak magnitude during each switching period. The logic controller 208 processes the modulation signal 226 and outputs a control signal 230 to the driving component 210 which generates the signal 156 to turn on or off the switch 128. For example, the demagnetization detector 212 detects the feedback signal 160 and outputs a signal 232 for determining the end of the demagnetization process of the secondary winding 114. In another example, the demagnetization detector 212 detects the feedback signal 160 and outputs the signal 232 for determining the beginning and the end of the demagnetization process of the secondary winding 114. In addition, the demagnetization detector 212 outputs a trigger signal 298 to the logic controller 208 to start a next cycle. The controller 102 is configured to keep an on-time period associated with the modulation signal 226 approximately constant for a given output load.
The controller 102 is operated in a voltage-mode where, for example, the signal 224 from the error amplifier 216 and the signal 228 from the oscillator 202 are both voltage signals and are compared by the comparator 206 to generate the modulation signal 226 to drive the power switch 128. Therefore, an on-time period associated with the power switch 128 is determined by the signal 224 and the signal 228.
FIG. 3 is a simplified conventional diagram showing the current-sensing component 214 and the error amplifier 216 as parts of the controller 102. The current-sensing component 214 includes a switch 302 and a capacitor 304. The error amplifier 216 includes switches 306 and 308, an operational transconductance amplifier (OTA) 310.
As shown in FIG. 3, the current-sensing component 214 samples the current-sensing signal 164 and the error amplifier 216 amplifies the difference between the signal 220 and the reference signal 222. Specifically, the switch 302 is closed (e.g., being turned on) or open (e.g., being turned off) in response to a signal 314 in order to sample peak magnitudes of the current-sensing signal 164 in different switching periods. If the switch 302 is closed (e.g., being turned on) in response to the signal 314 and the switch 306 is open (e.g., being turned off) in response to the signal 232 from the demagnetization detector 212, the capacitor 304 is charged and the signal 220 increases in magnitude. If the switch 306 is closed (e.g., being turned on) in response to the signal 232, the switch 308 is open (e.g., being turned off) in response to a signal 312 and the difference between the signal 220 and the reference signal 222 is amplified by the amplifier 310. The signal 312 and the signal 232 are complementary to each other. For example, during the demagnetization process of the secondary winding 114, the signal 232 is at a logic high level. The switch 306 remains closed (e.g., being turned on) and the switch 308 remains open (e.g., being turned off). The OTA 310, together with the capacitor 134, performs integration associated with the signal 220.
Under stable normal operations, an average output current is determined, according to the following equation, without taking into account any error current:
                                          I            o                    _                =                              1            2                    ×          N          ×                                    V              ref_ea                                      R              cs                                                          (                  Equation          ⁢                                          ⁢          1                )            where N represents a turns ratio between the primary winding 112 and the secondary winding 114, Vref_ea represents the reference signal 222 and Rcs represents the resistance of the resistor 130. As shown in Equation 1, the parameters associated with peripheral components, such as N and Rcs, can be properly selected through system design to achieve output current regulation.
For LED lighting, efficiency, power factor and total harmonic are also important. For example, efficiency is often needed to be as high as possible (e.g., >90%), and a power factor is often needed to be greater than 0.9. Moreover, total harmonic distortion is often needed to be as low as possible (e.g., <10%) for some applications. But the system 100 often cannot satisfy all these needs.
Hence it is highly desirable to improve the techniques of regulating output currents of power conversion systems.